Achint Soni

Actively seeking a PhD position starting in Fall 2025.

I am a Research Masters (Thesis) student in Computer Science at the Cheriton School of Computer Science, University of Waterloo, advised by Prof. Sirisha Rambhatla and Prof. Charles Clarke. Previously, I received my B.Tech in Electrical Engineering from the Indian Institute of Technology Kanpur, where I was advised by Prof. Faiz Hamid and Prof. Laxmidhar Behera.

My primary research areas are:

Foundation models (including generative world models), image and video generation/editing, multimodal (vision/language) understanding, and disentangled representation learning.

If you want to collaborate/have any questions feel free to shoot me an email. I am always interested in connecting with people.

News

[May'24] Our work "Why do Variational Autoencoders really promote disentanglement?" got accepted to ICML 2024.
[Sep'23] Joined David R. Cheriton School of Computer Science at University of Waterloo as a graduate student
[May'23] Graduated from IIT Kanpur with a B.Tech. degree majoring in Electrical Engineering, with minors in machine learning and theory of computation.
[Apr'23] received an offer of admission to the MMath in Computer Science program at the University of Waterloo.

Publications

VideoAgent: Self-Improving Video Generation

Achint Soni, Sreyas Venkataraman, Abhranil Chandra, Sebastian Fischmeister, Percy Liang, Bo Dai, Sherry Yang

Preprint, under review at ICLR 2025

VIDEOSCORE: Building Automatic Metrics to Simulate Fine-grained Human Feedback for Video Generation

Xuan He, Dongfu Jiang, Ge Zhang, Max Ku, Achint Soni, Sherman Siu, Haonan Chen, Abhranil Chandra, Ziyan Jiang, Aaran Arulraj, Kai Wang, Quy Duc Do, Yuansheng Ni, Bohan Lyu, Yaswanth Narsupalli, Rongqi Fan, Zhiheng Lyu, Yuchen Lin, Wenhu Chen

EMNLP Main, 2024

Why Do Variational Autoencoders Really Promote Disentanglement?

Pratik Bhowal, Achint Soni, Sirisha Rambhatla

International Conference on Machine Learning (ICML), 2024